Differential privacy (DP) is a prominent technique for protecting sensitive patient data in medical deep learning (DL), yet deploying it without compromising clinical utility or equity remains challenging. This scoping review synthesizes applications of DP in medical DL across centralized and federated settings. A structured search identified 74 eligible studies published through March 2025. Across modalities and tasks, DP, especially via DP-SGD, can maintain clinically acceptable performance under moderate privacy budgets (ϵ ≈ 10), particularly in imaging. However, strict privacy (ϵ ≈ 1) often leads to substantial accuracy loss, with amplified degradation in smaller or heterogeneous datasets. Only a minority of studies evaluate fairness, and several report that DP can widen subgroup performance gaps. Beyond DP-SGD, alternative mechanisms, including generative modeling, local DP, and hybrid federated designs, are emerging, but reporting of privacy parameters remains inconsistent. We identify key gaps in fairness auditing and standardization, and outline priorities for equitable, clinically robust privacy-preserving DL.
Bone fractures represent a significant global healthcare burden. Although fractures typically heal on their own, some fail to regenerate properly, leading to nonunion, a condition that causes prolonged disability, morbidity, and mortality. The challenge of treating nonunion fractures is further complicated in patients with underlying bone disorders where systemic and local factors impair bone healing. Traditional treatment approaches, including autografts, allografts, xenografts, and synthetic biomaterials, face limitations such as donor site pain, immune rejection, and insufficient mechanical strength, underscoring the need for alternative strategies. Biologic therapies have emerged as promising tools to enhance bone regeneration by leveraging the body’s natural healing processes. This review explores the critical role of conventional and emerging biologics in fracture healing. We categorize biologic therapies into protein-based treatments, gene and transcript therapies, small molecules, peptides, and cell-based therapies, highlighting their mechanisms of action, advantages, and clinical relevance. Finally, we examine the potential applications of biologics in treating fractures associated with bone disorders such as osteoporosis, osteogenesis imperfecta, rickets, osteomalacia, Paget’s disease, and bone tumors. By integrating biologic therapies with existing biomaterial-based strategies, these innovative approaches have the potential to transform clinical management and improve outcomes for patients with difficult-to-heal fractures.
Articular cartilage is crucial for joint function; however, it has limited regenerative capacity when damaged, a hallmark of many rheumatic diseases. Non-invasive imaging is essential for early diagnosis, therapeutic monitoring and prognostication. MRI remains the reference standard, offering detailed assessment of both morphological and compositional cartilage changes. Technological advances, including high-resolution and compositional MRI techniques such as T2 mapping, T1ρ, delayed gadolinium-enhanced MRI of cartilage, sodium imaging, diffusion imaging and ultra-short echo-time imaging, enable early detection of matrix alterations that precede structural breakdown. CT arthrography, although it involves radiation, serves as a valuable alternative when MRI is contra-indicated, offering high performance in the detection and evaluation of cartilage surface lesions. Emerging modalities, such as ultrasonography and PET, offer additional functional insights but are currently limited in scope. Artificial intelligence is poised to transform cartilage imaging through accelerated acquisition, automated segmentation, improved interpretation and enhanced efficiency, with growing clinical adoption. Advanced cartilage imaging will probably have an increasingly important role in clinical rheumatology, particularly for the optimization of individualized management of cartilage pathology. Non-invasive imaging of articular cartilage has evolved markedly and can be used to monitor response to treatment and predict disease outcomes. This Review provides rheumatologists with a comprehensive update on current and emerging imaging and analysis techniques for the assessment of cartilage.
Aktualisierungshinweis Die DDG-Praxisempfehlungen werden regelmäßig zur zweiten Jahreshälfte aktualisiert. Bitte stellen Sie sicher, dass Sie jeweils die neueste Version lesen und zitieren.
